Ocean init mode: add subroutine for smoothing topography #440
Conversation
... instead of "smooth_topography" because this is a more accurate description of what is done when this flag is .true.
|
Currently, I'm having trouble with scratch fields (at least on my laptop). I'll try on LANL IC tomorrow and ask @mark-petersen for help if I'm still not able to figure it out. |
xylar
force-pushed
the
ocean/add_topo_smoothing
branch
from
b5c9305 to
9accc29
Compare
This feature performs an iterative weighted average of a field (currently bottomDepthObserved) with a given smoothing weight and for a given number of iterations. Currently, it is only used for wISC setups.
xylar
force-pushed
the
ocean/add_topo_smoothing
branch
from
9accc29 to
3deec68
Compare
Previously, invalid cells were getting averaged in.
Results EC60to30wISCBelow are the bathymetry, ice draft and fraction of ice-shelf cover, without smoothing followed by with smoothing with a weight (of the central cell relative to the average of its neighbors) of 0.9 and 6 iterations.
|
|
weight of 0.9 and 6 iterations might be too much for this data set, particularly because it was already smoothed at 0.1 degrees on the original grid, but it does demonstrate the concept. I think this amount of smoothing might be more appropriate at higher res and in particular for the BedMachine bathymetry (unsmoothed) that we will hopefully move to for Cryosphere configurations. |
Results SO60to10wISCSame as above but at higher resolution.
In my testing, the SSH adjustment phase failed, so there's clearly some work still to do to make that more robust but I will do that in #433 with the new bathymetry. |
|
@mark-petersen, I assume you can handle the registry conflict but let me know. |
mark-petersen
left a comment
mark-petersen
left a comment
There was a problem hiding this comment.
Looks good by visual inspection. Passes nightly regression suite on grizzly, gnu/optimized, intel/debug. I did not look at the actual smoothed bathymetry results, but @xylar's plots above are sufficient.
…elop This feature performs an iterative weighted average of a field with a given smoothing weight and for a given number of iterations. Currently, it is only used for Cryosphere test cases (with ice-shelf cavities, wISC). Also renames config option ...fill_bathymetry_holes from ...smooth_topography because this is a more accurate description of what is done when this flag is .true. closes MPAS-Dev#436
…elop This feature performs an iterative weighted average of a field with a given smoothing weight and for a given number of iterations. Currently, it is only used for Cryosphere test cases (with ice-shelf cavities, wISC). Also renames config option ...fill_bathymetry_holes from ...smooth_topography because this is a more accurate description of what is done when this flag is .true. closes MPAS-Dev#436
…elop This feature performs an iterative weighted average of a field with a given smoothing weight and for a given number of iterations. Currently, it is only used for Cryosphere test cases (with ice-shelf cavities, wISC). Also renames config option ...fill_bathymetry_holes from ...smooth_topography because this is a more accurate description of what is done when this flag is .true. closes MPAS-Dev#436
Following the SORRM mesh from compass/legacy, we perform 10 iterations of smoothing with a weight of 0.92. For more info on the smoothing approach, see: MPAS-Dev/MPAS-Model#440 For more on the chosen values (related to ice-shelf cavities), see: MPAS-Dev/MPAS-Model#532 (comment)
Following the SORRM mesh from compass/legacy, we perform 10 iterations of smoothing with a weight of 0.92. For more info on the smoothing approach, see: MPAS-Dev/MPAS-Model#440 For more on the chosen values (related to ice-shelf cavities), see: MPAS-Dev/MPAS-Model#532 (comment)
We perform 6 iterations of smoothing with a weight of 0.9. For more info on the smoothing approach, see: MPAS-Dev/MPAS-Model#440 For more on previously chosen values (related to ice-shelf cavities), see: MPAS-Dev/MPAS-Model#532 (comment)
3 participants
This feature performs an iterative weighted average of a field with a given smoothing weight and for a given number of iterations. Currently, it is only used for Cryosphere test cases (with ice-shelf cavities,
wISC).Also renames config option
...fill_bathymetry_holesfrom...smooth_topographybecause this is a more accurate description of what is done when this flag is.true.closes #436